MLSP-L33: Multimodal Learning and Large Language Models
Oral
Fri, 8 May, 14:00 - 16:00
Location: Room 120+121
Session Type: Oral
Track: Machine Learning for Signal Processing [ML]
Click the to view the manuscript on IEEE Xplore Open Preview
Fri, 8 May, 14:00 - 14:20

MLSP-L33.1: DREAM FRAGMENT: INSTANCE-AWARE TEXT-TO-3D GENERATION FOR COMPOSITIONAL MULTI-OBJECT SCENES WITH COMPLEX INTERACTIONS
Zihao Yang, Chaofan Luo, University of Science and Technology of China, China; Conglang Zhang, Wuhan University, China; Xun Yang, University of Science and Technology of China, China; Xuyang Chen, Technical University of Munich, Germany
Fri, 8 May, 14:20 - 14:40

MLSP-L33.2: CONFCLIP: CONFIDENCE-WEIGHTED AND CLIPPED REWARD FOR REINFORCEMENT LEARNING IN LLMS
Bonan Zhang, Shanghai JiaoTong University, China; Zhongqi Chen, Bowen Song, Ant Group, China; Qinya Li, Fan Wu, Guihai Chen, Shanghai JiaoTong University, China
Fri, 8 May, 14:40 - 15:00

MLSP-L33.3: MAG: Multi-Modal Aligned Autoregressive Co-Speech Gesture Generation without Vector Quantization
Binjie Liu, Communication University of China, China; Lina Liu, China Mobile Research Institute, Beijing, China, China; Sanyi Zhang, Communication University of China, China; Songen Gu, Fudan University, China; Yihao Zhi, The Chinese University of Hong Kong(Shenzhen), China; Tianyi Zhu, Lei Yang, China Mobile Research Institute, Beijing, China, China; Long Ye, Communication University of China, China
Fri, 8 May, 15:00 - 15:20

MLSP-L33.4: NATIVETOK: NATIVE VISUAL TOKENIZATION FOR IMPROVED IMAGE GENERATION
Bin Wu, Mengqi Huang, Weinan Jia, Zhendong Mao, University of Science and Technology of China, China
Fri, 8 May, 15:20 - 15:40

MLSP-L33.5: Disentangled Signals, Dynamic Prompts: A Meta-Network Framework for Robust Task-Oriented Dialogue
Changzhu Xiong, Yuexian Zou, Peking University, China; Tianyu Li, Jilin Institute of Chemical Technology, China
Fri, 8 May, 15:40 - 16:00

MLSP-L33.6: DO SPEECH LLMS LEARN CROSSMODAL EMBEDDING SPACES?
Carlos Escolano, Universitat Politècnica de Catalunya (UPC), Spain; Gerard Sant, University of Zurich, Switzerland; José A.R. Fonollosa, Universitat Politècnica de Catalunya (UPC), Spain